ArXiv Preprint
Semi-supervised learning (semi-SL) is a promising alternative to supervised
learning for medical image analysis when obtaining good quality supervision for
medical imaging is difficult. However, semi-SL assumes that the underlying
distribution of unaudited data matches that of the few labeled samples, which
is often violated in practical settings, particularly in medical images. The
presence of out-of-distribution (OOD) samples in the unlabeled training pool of
semi-SL is inevitable and can reduce the efficiency of the algorithm. Common
preprocessing methods to filter out outlier samples may not be suitable for
medical images that involve a wide range of anatomical structures and rare
morphologies. In this paper, we propose a novel pipeline for addressing
open-set supervised learning challenges in digital histology images. Our
pipeline efficiently estimates an OOD score for each unlabelled data point
based on self-supervised learning to calibrate the knowledge needed for a
subsequent semi-SL framework. The outlier score derived from the OOD detector
is used to modulate sample selection for the subsequent semi-SL stage, ensuring
that samples conforming to the distribution of the few labeled samples are more
frequently exposed to the subsequent semi-SL framework. Our framework is
compatible with any semi-SL framework, and we base our experiments on the
popular Mixmatch semi-SL framework. We conduct extensive studies on two digital
pathology datasets, Kather colorectal histology dataset and a dataset derived
from TCGA-BRCA whole slide images, and establish the effectiveness of our
method by comparing with popular methods and frameworks in semi-SL algorithms
through various experiments.
Nikhil Cherian Kurian, Varsha S, Abhijit Patil, Shashikant Khade, Amit Sethi
2023-03-17
